Pelagic Hyperiidea — Bary 
325 
mopolites may occur at stations located in a 
variety of waters. Thus, Phronima sedentaria 
(Forskal) is present in water of subtropical 
origin north of the subtropical convergence 
(Station 326, Figs. 2, 6d), in mixed water 
(Station 79, Fig. 5^), and in water predomi- 
nantly of subantarctic origin (Stations 214, 
Fig. 5g), or being influenced by this water 
(Station 279, Fig. 5/6). Hyperia hengalensis 
(Giles) was present in mixed or diluted coastal 
waters (Stations 79, 100), and Primno macropa 
Guer., another cosmopolitan species, also oc- 
curred at Station 79 (Figs. 2, '5a). Thus, on the 
whole, these species are shown to have been 
captured in conditions suitable for them. 
Conversely, their occurrences in conditions 
consistent with those previously recorded for 
them adds to the value of interpretations 
based on the more commonly occurring 
species. 
Two of the New Zealand species listed in 
Table 1 are regarded as being doubtfully 
identified (Hurley, 1955). The identification 
of one, and possibly also of the other species, 
is not upheld when their relationships to the 
water masses, demonstrated in the T-S-P di- 
agram, are compared with their previously 
recorded distributions. Vibilia stebbingi (?) 
Behn and Wolt, is present in water of sub- 
antarctic origin (Station 210, Figs. 2, 5g) 
which is out of character with the tropical- 
subtropical range usually ascribed to this 
species. Parascelis typhoides (.^) Claus, another 
tropical-subtropical species was captured at 
Station 189 (Figs. 2, 5/), believed to be lo- 
cated in a mixture of coastal and subantarctic 
waters; the absence of any others of the se- 
lected Subtropical Group of species indicates 
little influence at this point from water of 
subtropical origin. Thus this occurrence of P. 
typhoides (.^) may also be anomalous, suggest- 
ing again that misidentification is possible. 
A comparison of the average numbers of a 
species captured in coastal and in offshore 
waters may indicate the degree to which the 
species penetrate from one water into the 
other. When numbers captured per haul over 
a specified range of depth are plotted against 
depth of water, the commoner amphipod 
species illustrate that this applies. (See Fig. 
7; deeper water in the area of sampling is 
indicative of an increase of distance offshore 
and proximity to oceanic water.) 
Both P. gaudichaudii and Cyllopus magel- 
lanicus (Fig. 8, unbroken line) show de- 
cided increases in the numbers captured from 
shallow (coastal) to deep water. A fair degree 
of tolerance to coastal waters is suggested for 
P. gaudichaudii by the almost steady increase 
in numbers as samples proceed offshore. On 
the other hand, the sudden decrease in water 
shallower than 50 fathoms (91.5 m.) shown by 
C, magellanicus is indicative of intolerance to 
conditions in the coastal water. P. gracilipes 
and P. australis (Fig. 7) increase in numbers 
to a peak at 50 fathoms (91.5 m.), and de- 
crease in the deeper, offshore water. This sug- 
gests that neither species is tolerant of 
conditions in this water (see later). These 
facts confirm the relationships of the species 
to water masses already obtained from the 
T-S-P diagrams (Figs. 2, 3, 4). 
The collections of this study were made 
during a period of three months. During the 
latter half of this a larger proportion of sta- 
tions were over deeper water, either directly 
influenced by, or believed to be situated in 
water of subantarctic origin. It seemed pos- 
sible, therefore, that variations in the catches 
of a species, relative to depth of water, might 
be reflected in the numbers captured at differ- 
ent times in the three months. So that direct 
comparisons might be made, the average 
number of specimens per haul for each month 
are included in the figures of changes of the 
catch with depth (Fig. 7 dashed line). Cyl- 
lopus magellanicus shows an overall decrease in 
numbers during January through March 
which is opposed to the increase with depth. 
Numbers of P. gaudichaudii on the other 
hand, increase with each monthly catch, al- 
though less so for March than for January 
and February. Even so, the number collected 
